Olson / Wu Predictive Data Mining Models

2. Auflage 2020
ISBN: 978-981-13-9664-9
Verlag: Springer Nature Singapore
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 131 Seiten

Reihe: Computational Risk Management

This book provides an overview of predictive methods demonstrated by open source software modeling with Rattle (R') and WEKA. Knowledge management involves application of human knowledge (epistemology) with the technological advances of our current society (computer systems) and big data, both in terms of collecting data and in analyzing it. We see three types of analytic tools. Descriptive analytics focus on reports of what has happened. Predictive analytics extend statistical and/or artificial intelligence to provide forecasting capability. It also includes classification modeling. Prescriptive analytics applies quantitative models to optimize systems, or at least to identify improved systems.  Data mining includes descriptive and predictive modeling.  Operations research includes all three. This book focuses on prescriptive analytics.
The book seeks to provide simple explanations and demonstration of some descriptive tools. This second edition provides more examples of big data impact, updates the content on visualization, clarifies some points, and expands coverage of association rules and cluster analysis. Chapter 1 gives an overview in the context of knowledge management. Chapter 2 discusses some basic data types. Chapter 3 covers fundamentals time series modeling tools, and Chapter 4 provides demonstration of multiple regression modeling. Chapter 5 demonstrates regression tree modeling.  Chapter 6 presents autoregressive/integrated/moving average models, as well as GARCH models. Chapter 7 covers the set of data mining tools used in classification, to include special variants support vector machines, random forests, and boosting.  
Models are demonstrated using business related data. The style of the book is intended to be descriptive, seeking to explain how methods work, with some citations, but without deep scholarly reference. The data sets and software are all selected for widespread availability and access by any reader with computer links.


David L. Olson is the James & H.K. Stuart Chancellor's Distinguished Chair and Full Professor at the University of Nebraska.  He has published research in over 150 refereed journal articles, primarily on the topic of multiple-objective decision-making, information technology, supply chain risk management, and data mining.  He teaches in the management information systems, management science, and operations management areas.  He has authored over 20 books and is a member of the Decision Sciences Institute, the Institute for Operations Research and Management Sciences, and the Multiple Criteria Decision Making Society.  He was a Lowry Mays endowed Professor at Texas A&M University from 1999 to 2001.  He was named the Raymond E. Miles Distinguished Scholar for 2002, and was a James C. and Rhonda Seacrest Fellow from 2005 to 2006.  He was named Best Enterprise Information Systems Educator by the IFIP in 2006 and is a Fellow of the Decision Sciences Institute.
Desheng Dash Wu is a distinguished professor at the University of Chinese Academy of Sciences. His research interests include enterprise risk management, performance evaluation, and decision support systems. His has published more than 80 journal papers in such journals as Production and Operations Management, IEEE Transactions on Systems, Man, and Cybernetics: Systems, Risk Analysis, Decision Sciences, Decision Support Systems, European Journal of Operational Research, IEEE Transactions on Knowledge and Data Engineering, et al. He has coauthored 3 books with David L Olson, and has served as editor/guest editor for several journals such as IEEE Transactions on Systems, Man, and Cybernetics: Part B, Omega, Computers and OR, International Journal of Production Research.

Autoren/Hrsg.

Olson, David L.

Wu, Desheng

Inhaltsverzeichnis

1;Preface;6
2;Contents;7
3;About the Authors;10
4;1 Knowledge Management;11
4.1;1.1 The Big Data Era;11
4.2;1.2 Business Intelligence;12
4.3;1.3 Knowledge Management;13
4.4;1.4 Computer Support Systems;13
4.5;1.5 Data Mining Forecasting Applications;15
4.6;1.6 Data Mining Tools;16
4.7;1.7 Summary;17
4.8;References;18
5;2 Data Sets;20
5.1;2.1 Gold;21
5.2;2.2 Other Datasets;23
5.2.1;2.2.1 Financial Index Data;23
5.2.2;2.2.2 Loan Analysis Data;26
5.2.3;2.2.3 Job Application Data;26
5.2.4;2.2.4 Insurance Fraud Data;27
5.2.5;2.2.5 Expenditure Data;27
5.3;2.3 Summary;28
5.4;References;29
6;3 Basic Forecasting Tools;30
6.1;3.1 Moving Average Models;30
6.2;3.2 Regression Models;31
6.3;3.3 Time Series Error Metrics;35
6.4;3.4 Seasonality;36
6.5;3.5 Demonstration Data;40
6.6;3.6 Software Demonstrations;44
6.6.1;3.6.1 R Software;45
6.6.2;3.6.2 Weka;48
6.7;3.7 Summary;51
7;4 Multiple Regression;54
7.1;4.1 Data Series;54
7.2;4.2 Correlation;56
7.3;4.3 Lags;59
7.4;4.4 Summary;64
8;5 Regression Tree Models;66
8.1;5.1 R Regression Trees;66
8.2;5.2 Random Forests;70
8.3;5.3 WEKA Regression Trees;76
8.3.1;5.3.1 Decision Stump;77
8.3.2;5.3.2 Random Tree Modeling;78
8.3.3;5.3.3 REP Tree Modeling;79
8.3.4;5.3.4 Random Forest;79
8.3.5;5.3.5 M5P Modeling;85
8.4;5.4 Summary;86
8.5;Reference;86
9;6 Autoregressive Models;87
9.1;6.1 ARIMA Models;87
9.2;6.2 ARIMA Model of Brent Crude;88
9.3;6.3 ARMA;90
9.4;6.4 GARCH Models;95
9.4.1;6.4.1 ARCH(q);95
9.4.2;6.4.2 GARCH(p, q);95
9.4.3;6.4.3 EGARCH;96
9.4.4;6.4.4 GJR(p, q);96
9.5;6.5 Regime Switching Models;97
9.6;6.6 Application on Crude Oil Data;97
9.7;6.7 Summary;100
9.8;References;101
10;7 Classification Tools;102
10.1;7.1 Bankruptcy Data Set;102
10.2;7.2 Logistic Regression;104
10.3;7.3 Support Vector Machines;109
10.4;7.4 Neural Networks;113
10.5;7.5 Decision Trees;114
10.6;7.6 Random Forests;117
10.7;7.7 Boosting;119
10.8;7.8 Comparison;120
10.9;7.9 WEKA Classification Modeling;123
10.10;7.10 Summary;127
10.11;Reference;128
11;8 Predictive Models and Big Data;129
11.1;References;131